1. Field of the Invention
This invention relates to a multi-channel recording apparatus in which simultaneous recording of signals of a plurality of channels is made on a tape-shaped recording medium (hereinafter simply called "tape").
2. Description of the Related Art
Recently, the electronic technique, high-precision machining technique and the like have advanced to improve the magnetic recording density and a digital signal processing technique has been developed. These lead, at present, to realize digital magnetic recording of sound signals. Further, even in the field of video signals, application of the digital recording technique is being promoted. However, because the digital signal necessarily has as wide a band as more than ten times that of the analog signal, to allow for the magnetic recording and/or reproduction apparatus, for example, VTR, to record the digital video signal, either one of the following means must be carried out: in comparison with the case of the analog video signals, (i) to raise the relative speed of the magnetic tape and the magnetic head, (ii) to divide the video signal into a plurality of channels, and (iii) to increase the recording density by altering the materials of the magnetic tape and the magnetic head. Of these, (ii) has the following merits over the others:
1. The transmission band can be lowered to 1/N (where N is the number of channels).
2. The mechanical problems such as head touch, tape damage or the like and the problem of fidelity are fewer.
In the past, as the head arrangement for a VTR in which the video signal is divided into a plurality of channels as has been described above, and these are simultaneously recorded on, or reproduced from, the magnetic tape or the like, what is depicted in FIG. 1 has been considered. For note, the illustration is made for a case of three channels, or N=3.
As shown in FIG. 1, a rotary drum 1 of cylindrical shape has 2N or 6 magnetic heads 2A, 2B, 3A, 3B, 4A and 4B arranged in 60.degree. spaced relation to each other. Also, the paired magnetic heads 2A and 2B, 3A and 3B, or 4A and 48 are differentiated 180.degree. in phase from each other. The three flows of signals are allocated to the respective individual pairs of magnetic heads, so that they are simultaneously recorded on the magnetic tape T is trained round the rotary drum 1 over an angular range of more than 180.degree. by loading posts 5 and 5'. FIG. 2 illustrates recording patterns on the tape as the head arrangement of FIG. 1 operates. In this case, by changing the azimuth angle between the magnetic heads denoted by the subscripts A and B in each pair, the azimuth angle of the adjacent tracks formed on the magnetic tape T can be changed. This makes it possible to achieve a high guard-bandless recording density.
In such a manner, according to the above-described apparatus, the signal which when digitized would otherwise take a very wide band such as the video signal also can be recorded with reduction of that band to 1/N (in this instance, 1/3). None the less, in this apparatus, it is also possible to decrease the tape feeding speed to 1/3 and use one pair of the, magnetic heads 2A and 2B to record only one flow of signals. If the diameter of the rotary drum 1, the arrangement of the magnetic heads and others are made set in conformity with the already existing single-channel type apparatus, therefore, compatibility can be preserved therebetween. Hence, the multi-channel apparatus is evaluated as having excellent utilities.
However, in the magnetic recording-reproduction apparatus of the configuration described above, in order to make simultaneous recording of N channels, 2N magnetic heads must be used. Moreover, the equal number of transmission systems such as rotary transmitters, pre-amplifiers and others to the number of used magnetic heads must be provided. Therefore, a problem arose that the apparatus increased in the entire scale, and became expensive.
Also, it is being considered that the number of magnetic heads to be installed on the rotary drum is made equal to the number of divided channels and the angle the magnetic tape is trained round this drum is increased nearly to 360.degree., when the recording or reproducing is performed in one channel by each head. In this case, because the signal drops out for a prescribed period during the time when the drum 1 rotates one revolution, a time axis compression circuit and either a circuit for interpolating that period or a time axis expanding circuit must be provided. Therefore, an alternate problem arises that the complexity of structure of the signal processing system is increased. Further, as the diameter of the rotary drum is made smaller, the difficulty of training the magnetic tape on it increases, because the training angle is as wide as 360.degree.. This also causes the dropping-out period of the signal to increase. Therefore, an additional problem arises that the apparatus cannot be minimized in size.
By the way, the conventional single-channel magnetic recording and reproduction apparatus necessarily has the capability of automatic tracking control to operate when in the reproducing mode. With this, the tracking error of the reproducing head is automatically corrected. For this purpose, there have been known two methods in the art, one of which is to use a control signal recorded in the linear track in correcting the tracking error of the reproducing head (CTL method), and the other of which is first to record four pilot signals of low frequencies slightly differing from each other successively one at an allocated track thereto, and, when to reproduce, then to use the cross talk reproduced from the adjacent tracks by the magnetic heads in correcting the tracking error (pilot method).
However, in application of such automatic tracking methods as has been described above to the recording and reproduction apparatus capable of selectively recording one flow of signals and three flows of signals, there were the following problems.
In the case of, for example, the CTL method, when one flow of signals is recorded and reproduced, the CTL signal takes the form of one pulse signal per two tracks or one track, as is well known in the art. But, if the number of the pulse signal per two tracks or one track is 1 even for the recording of the three flows of signals, no discrimination from the area in which one flow of signals was recorded can be made when in the reproduction mode. Therefore, when that area of the tape which has the signals of three channels recorded thereon is to be reproduced, it will result that the tape is caused to run at the same speed when the signal of single channel is reproduced. This implies that the reproducing is operated in the same manner. Thus, a problem arises that there are occasions of failing reproduction of the three flows of signals. To avoid this, it may be considered that when the recording is changed over from the one flow to the three flows of signals, the CTL signal is made to correspond to one of the three flows so that when in reproduction mode, it becomes possible to make discrimination between the single-channel and triple-channel recorded areas of the tape. But if it happens that the tape running speed when in the single-channel recording becomes equal to that when in the triple-channel recording, in other words, the track pitch for the former differs 3 times as wide from that for the latter, the above-described discrimination cannot also be made. Also, the use of the CLT method necessitates a track and a magnetic head both solely used for automatic tracking. This hinders a minimization of the size of the apparatus and an increase of the recording density from being achieved.
Meanwhile, as the pilot method is applied, if the four pilot signals f.sub.1 to f.sub.4 are recorded successively on the respective tracks TR.sub.1 to TR.sub.4 in superimposed relation to the video signals as shown in FIG. 3, the triple-channel and single-channel recorded areas cannot be discriminated from each other. And there is need to differentiate all these pilot signals in frequency and also the timings at which they are produced from one another, complicating the structure of the required circuit therefor. To avoid this, if the succession of the four pilot signals is adjusted according to one of the three flows of signals as shown in FIG. 4, the above-identified discrimination becomes possible to perform. But, because there are a good many of those sections of the boundaries between the adjacent two tracks across which the frequencies of the pilot signals come to be the same, it will become impossible in between the adjacent tracks at those sections to discriminate which track should be sought. Hence, the tracking control can no longer be made. Of course, even for this case, if a sample and hold circuit is used, the tracking error can be detected. In more detail, as, for example, the third head errs from tracking the third track TR.sub.3, when it is sided toward the second track TR.sub.2, its output includes the pilot signal f.sub.4. The level of this pilot signal after having been sampled and held is compared with the level of the pilot signal f.sub.2 obtained from the adjacent track TR.sub.4 to the track TR.sub.3, as the latter is also sampled and held, when that error of tracking is detected. If the lineality of the track is bad, however, such a measure cannot provide a sufficient tracking performance. Another problem is that the scale of circuitry is very increased.